This proposal is for support of research to explore a biochemical method of cancer detection based on quantitation of serum levels of a heteroglycan fraction of undetermined composition which is consistently and markedly elevated in tumor-bearing animal models. Specific objectives include the following: 1) To maximize reproducibility and sensitivity of the colorimetric assay procedure for the tumor-elevated heteroglycan fraction of the serum. 2) To correlate heteroglycan levels with tumorigenesis in a broad range of animal models to gain information on the generality of the heteroglycan change as a diagnostic tool. 3) To isolate and identify chemically the heteroglycan components of the serum which are elevated as a result of a tumor burden so that a more specific assay may be developed as well as to provide basic information necessary to understand the underlying biological mechanisms. Especially in the 02 and 03 years, the study will be expanded to include evaluations of potential clinical applications of the detection method in studies with sera of cancer patients. Serum from tumor-bearing individuals will be analyzed and compared to age and strain matched controls. The heteroglycan fraction will be prepared by a simplified extraction and purification procedure and quantitated on the basis of total sialic acid per mg protein or per ml serum. Results can be obtained with as little as 0.5 ml of whole blood. Initial emphasis will be on detailed assessment of the range of natural variation in the parameters monitored, influence of physiological, environmental, and nutritional factors on the heteroglycan fraction in normal animals, and procedural modifications to improve sensitivity and reliability. This will be followed by a survey of a variety of tumor-bearing animals, including man, to determine generality with a final view toward clinical applications.